Available online at ScienceDirect. Energy Procedia 63 (2014 ) GHGT-12

Transcription

1 Available online at ScienceDirect Energy Procedia 63 (2014 ) GHGT-12 A workflow to determine CO 2 storage potential in deep saline formations Wesley A. Peck a *, Kyle A. Glazewski a, Robert C.L. Klenner a, Charles D. Gorecki a, Edward N. Steadman a, and John A. Harju a a Energy & Enviromental Research Center, 15 North 23rd Street, Stop 9018, Grand Forks, ND , United States Abstract A workflow was developed to properly assess the CO 2 storage resource potential of a deep saline formation using the methodology proposed by the U.S. Department of Energy. To illustrate the workflow, the Minnelusa Formation of the Powder River Basin is used as an example of how a CO 2 storage resource methodology could be applied to a saline formation given varying levels of information. It is important to accurately estimate the effective volumetric CO 2 storage resource potential of a target formation, and new storage efficiency values are presented with the workflow Energy The Authors. & Environmental Published Research by Elsevier Center, Ltd. University of North Dakota. Published by Elsevier Limited. This is an open access article under the CC BY-NC-ND license ( Selection and peer-review under responsibility of GHGT. Peer-review under responsibility of the Organizing Committee of GHGT-12 Keywords: storage capacity, saline formations, methodology, CO 2 storage. 1. Introduction Because of the concern over the growing amount of anthropogenic CO 2 emissions, using carbon capture and storage (CCS) to geologically store CO 2 is a key mitigation technique under consideration. The large volumes of CO 2 that would need to be stored in order to make a significant reduction in CO 2 emissions require the ability to have an accurate understanding of the CO 2 storage resource available in deep saline formations. The U.S. Department of Energy (DOE) established the Regional Carbon Sequestration Partnerships (RCSP) in 2003 to promote CCS in different regions of the United States and Canada and to determine and implement the technology, infrastructure, and regulations most appropriate to advance CCS. Part of this initiative included a characterization *Corresponding author Tel.: Tel.: ; fax: address: wpeck@undeerc.org Energy & Environmental Research Center, University of North Dakota. Published by Elsevier Limited. This is an open access article under the CC BY-NC-ND license ( Peer-review under responsibility of the Organizing Committee of GHGT-12 doi: /j.egypro

2 5232 Wesley A. Peck et al. / Energy Procedia 63 ( 2014 ) effort to determine the storage resource potential of saline formations in order to support future CCS demonstration projects. To date, these efforts have determined that the United States and Canada have an estimated CO 2 storage resource potential of 2012 to 20,043 billion tonnes in deep saline formations [1]. Characterization efforts are ongoing, and estimates continue to be revised and refined as more data become available. The interest in the geologic storage of CO 2 reinforces the importance of being able to accurately estimate the CO 2 storage resource potential of a given target formation. Conceptually, calculating the volumetric CO 2 storage potential of a given formation is a straight-forward task; however, each formation evaluation is unique and can differ significantly in terms of scope, budget, and available data. Basic information needed to assess a formation for CO 2 storage potential includes a formation s area, thickness, porosity, CO 2 density at reservoir conditions, and primary reservoir lithology. Other information such as salinity and distributions of porosity/permeability can further enhance the CO 2 storage estimate, if they are known. Depending on the quantity of basic data used to evaluate a saline formation, differing storage efficiency factors and their associated confidence intervals need to be applied to ensure the most accurate prediction possible. Lithology, depth, boundary conditions, and salinity all have a large impact on a formation s suitability for CO 2 storage. These parameters are also important in determining the portion of the formation that is amenable to CO 2 injection and storage, (i.e., the net-to-gross formation pore volume or E Geol [2]), and in choosing the appropriate storage efficiency term. The goal of this paper is to provide a step-by-step workflow to properly assess the CO 2 storage resource of a saline formation given varying levels of starting data. The workflow, terms, and concepts presented in this paper will provide the user confidence in performing CO 2 storage resource assessments, and help reduce under- or overestimation of the effective CO 2 storage resource potential of a target deep saline formation. For the purposes of this paper, the DOE National Energy Technology Laboratory (NETL) method was used [3]; however, a similar approach could be used for other volumetric approaches (e.g., Carbon Sequestration Leadership Forum, 2005 [4]). To help illustrate the workflow, three scenarios for the Minnelusa Formation of the Powder River Basin are presented. These examples demonstrate how a CO 2 storage resource estimate should be made for a saline formation with different levels of known information. 2. Methodology The two most commonly used methodologies to determine storage are those developed by DOE NETL [1] and the Carbon Sequestration Leadership Forum (CSLF) [4]. These two methodologies have been compared and found to be equivalent provided the same assumptions are made and the efficiency terms properly applied [2]. The effective CO 2 storage resource potential of a targeted saline formation is estimated using a volumetric equation where the pore volume of the target formation is multiplied by a storage efficiency term and the density of the CO 2 at reservoir conditions: = (1) To pore space available. The value for CO 2 2 to mass. The storage efficiency factor (E) represents the fraction of the total pore volume that can be occupied by the injected CO 2. Although volumetric methods such as this are straightforward, misapplications of the efficiency factors commonly occur and may ultimately lead to under- or overestimation of the effective storage resource potential of the formation under investigation. The choice of which efficiency terms to apply is directly related to the amount of information known about the formation s area, thickness, porosity, and pressure boundary conditions. The efficiency term (E) represents the percentage of the formation s pore volume that can be occupied by CO 2. In open systems, the efficiency term represents the fraction of the geology that is amenable to storage and the portion of that pore space that CO 2 can occupy by displacing the original formation fluids during the course of injection (E E ) (Equation 2). The amenable geology is defined as the fraction of the total formation volume that has suitable geology for CO 2 storage (E geol ) and is a multiplicative combination of the net-to-total area ( / ), the netto-gross thickness ( / ), and the effective-to-total porosity ( / ) (Equation 3). E geol is generally defined as the area where there is sufficient formation at a depth where CO 2 will remain in the supercritical state, typically around 800 meters, and where the salinity of the formation fluids is above the total dissolved solids (TDS) cutoff for

3 Wesley A. Peck et al. / Energy Procedia 63 ( 2014 ) protected underground sources of drinking water (USDW) (10,000 ppm). It also excludes intervals in the formation with unsuitable geology for injection. The second factor contained in the E E, the displacement efficiency (E D ), is split into the volumetric displacement efficiency (E vol ) and the microscopic displacement efficiency (E d ). The volumetric displacement efficiency is the combined fraction of the pore volume that can be contacted by CO 2 from injection wells and the fraction of the net thickness that is contacted by CO 2 from injection wells and the fraction of the net thickness that is contacted by CO 2 as a result of the density difference between the injected CO 2 and the formation fluids. The microscopic displacement efficiency represents the fraction of the contacted pore space that can be filled by CO 2 and is directly related to the irreducible water saturation. = (2) = / / / (3) = (4) To assist in identifying which efficiency factor to use for the volumetric calculation, a workflow was created to guide users in correctly assessing the formation under investigation (Fig.1). To illustrate the workflow, the upper Minnelusa Formation of the Powder River Basin is used as an example of how an effective CO 2 storage resource estimate could be determined from a saline formation with different levels of information. A 3-D geologic model of this formation was constructed for a study conducted for the IEA Greenhouse Gas R&D Programme (IEAGHG) [5]. The purpose of that study was to compare CO 2 storage efficiency in deep saline formations estimated using both volumetric and dynamic methodologies. The data from the IEAGHG study provide the baseline data for the scenarios presented in this paper. Three scenarios are presented: 1) if only the gross formation properties are known, 2) if all of the net-to-gross properties of the formation are known, or 3) if only the area amenable to CO 2 storage is known (net-to-gross area) property is known, with the other net-to-gross properties unknown. These are common scenarios that occur in CO 2 storage calculations. Those investigating potential target formations may know very little about the formation (Scenario 1), or they may be investing a large amount of time with abundant information known (Scenario 2). These two scenarios are straightforward in terms of procedure and which storage efficiency factor should be used. However, it is also common for investigators to have information on the formation s depth and salinity, which allows them to derive the net area (Scenario 3). In this case, the more conservative storage efficiency factor shouldn t be used because the net-to-gross factor is known, but there is not enough information known to apply the larger efficiency factors. As noted in Ellett and others [6], a simple disaggregation of the individual efficiency terms is not accurate. Therefore, the new efficiency factors shown in the workflow (Fig. 1) should be used. 3. Scenarios As described above, each scenario is presented using the upper Minnelusa Formation of the Powder River Basin. The upper Minnelusa Formation is a clastic, open boundary formation, and each scenario assumes a different combination of information is known through literature review, data gathering, or geologic modeling efforts. In order to inject CO 2 into a formation, the salinity must be above 10,000 ppm and the depth must be below 800 meters. If these two parameters are not met, the formation is not eligible (suitable) to be a storage target. If the reservoir meets these two requirements, the formation s total area, gross thickness, average porosity, and density of CO 2 at reservoir conditions must be determined (as shown in Fig. 1). Each scenario s parameter values used to calculate the CO 2 storage potential in the upper Minnelusa Formation are found in Tables 1 3. Table 1 shows the parameters that are used to calculate the total pore volume. For Scenarios 1 and 3, this involves simply multiplying the thickness, porosity, and area. Scenario 2 has values derived from a 3-D geocellular model, and pore volumes are determined through the model for each individual cell across the model.

4 5234 Wesley A. Peck et al. / Energy Procedia 63 ( 2014 ) Fig. 1. A workflow to estimate CO 2 storage resource in deep saline formations.

5 Wesley A. Peck et al. / Energy Procedia 63 ( 2014 ) There are key differences to note between the three scenarios. Scenario 2 has a lower average porosity value (3.35%) that was derived through the geocellular model. This value includes parts of the formation that are not amenable to CO 2 storage, whereas the 12% porosity value used in Scenarios 1 and 3 reflect only the part of the formation that is amenable to storage. The low porosity value in Scenario 2 results in a significantly lower total pore volume. Scenario 3 involves a notable difference in the knowledge regarding area of the formation. In Scenarios 1and 2, the Minnelusa Formation total area is known, regardless of whether the formation is deep enough or the salinity is high enough. In Scenario 3, only the net (target) area where the formation is deep enough and the salinity is high enough is analyzed. In other words, the total area of the formation is not used, and only the area meeting the CO 2 injection requirements is used to calculate pore volume. Table 2 shows the net-to-gross parameters that are known for each scenario. In Scenario 2, all the ratios for the key parameters are known which allows for the calculation of E geol and, consequently, higher E saline values to be used. In Scenarios 1 and 3, not all of the net-to-gross parameters are known, so E geol cannot be calculated. However, Scenario 3 does have the net-to-gross area as described above, so the total pore volume is equal to the net pore volume and a slightly higher E saline value (Table 3, Fig. 1) can be used to reduce the amount of pore space used for CO 2 storage. This E saline value in Scenario 3 accounts for the unknown net-to-gross thickness and porosity values. Table 3 shows the E saline values, CO 2 density, and the CO 2 storage values. Again, Scenario 2 has values derived from a 3-D geocellular model, and calculations are completed for individual cells across the model area Scenario 1 Scenario 1 represents the most basic approach to assessing the CO 2 storage of a formation. A formation being evaluated may have little information available, or those performing the assessment may have little time or resources to fully evaluate the formation. In order to estimate the CO 2 storage, the formation s average thickness, average porosity, and extent should be known. For Scenario 1, these values are determined through a literature review (Table 1). The full extent of the Minnelusa is not known in this scenario, and the literature review indicates that the formation is roughly the same area as the Powder River Basin at 51,800 km 2. Since the net-to-gross data are unknown, the most conservative low, mid, and high case values (P10, P50, P90) for E saline in a clastic formation must be used (Table 3, Fig. 1). Further, literature review found pressure and temperature values for the upper Minnelusa, resulting in an estimated average CO 2 density. The CO 2 storage in this scenario ranged from 1505 to 15,936 million tonnes (Table 3) Scenario 2 Scenario 2 represents a more thorough assessment where, unlike Scenario 1, all of the parameters are known. The formation is well understood, allowing for a more rigorous formation evaluation and a more accurate CO 2 storage estimate. Typically, in order to determine the necessary net-to-gross terms, a characterization effort including construction of a 3-D model is used. The 3-D model is constructed from available data sets and literature. Unlike Scenario 1, the formation extent and thickness were mapped by picking log tops to create detailed structure maps throughout the basin. A petrophysical workflow was performed to determine how porosity changes because of compaction and facies variations. The petrophysical results were distributed geostatistically to create a 3-D model of the formation capturing the overall heterogeneity. In contrast with Scenario 1, the net values for area, thickness, and porosity can be determined. The net area for the upper Minnelusa excludes areas where water salinity is <10,000 TDS and the measured depth is <800 meters. Net thickness excludes low porosity shales or tight dolomite where CO 2 could not be stored. Net porosity includes a cutoff based on a permeability/porosity cross-plot. Knowing these parameters allows for using higher efficiency factors than were used in Scenario 1 (Tables 2 and 3). The CO 2 storage in this scenario ranged from 4506 to 14,615 million tonnes (Table 3).

6 5236 Wesley A. Peck et al. / Energy Procedia 63 ( 2014 ) Scenario 3 Similar to Scenario 2, the formation s areal extent and thickness was mapped by picking log tops to create detailed structure maps throughout the basin. In Scenario 3, the part of the formation amenable to CO 2 injection (i.e., target area) is the only area used in the calculation, while the thickness and porosity net-to-gross parameters are unknown. Unlike Scenario 2, porosity is assumed from the literature review to be 12% as in Scenario 1. Since the target area is known, the formation s total area is equal to the net area that is known to have salinity greater than 10,000 TDS and is deeper than 800 meters (Table 1). This means the net pore volume is equal to the total pore volume, and the volume can apply a revised efficiency value that accounts for the missing net-to-gross thickness and porosity parameters (Table 3, Fig. 1). The estimated average CO 2 density used was the same as Scenario 1. The CO 2 storage in this scenario ranged from 5173 to 30,430 million tonnes (Table 3). Table 1. Scenario parameters used to calculate total pore space. Parameter Scenario 1 Scenario 2 Scenario 3 Average Thickness, m Average Porosity, % Total Area, km 2 51,800 70,300 58,350 Total Pore Volume, km Table 2. Scenario net-to-gross parameters used to determine efficiency values (E geol and E saline) used. Net-to-Gross Parameters Scenario 1 Scenario 2 Scenario 3 Area, % Unknown 0.87* 1 Thickness,% Unknown 0.84 Unknown Porosity, % Unknown 0.62 Unknown Net Pore Volume, km *Percentage in the model was volume, not area. Table 3. Scenario efficiency, density, and storage values. Parameter Scenario 1 Scenario 2 Scenario 3 E Saline Values, P10, P50, P90, %* 0.51, 2.0, , 14, , 4.41, 9.53 Average CO 2 Density, kg/m P10 CO 2 Storage, million tonnes P50 CO 2 Storage, million tonnes ,081 P90 CO 2 Storage, million tonnes 15,936 14,615 30,430 *P10, P50, P90 values given in order. 4. Summary As shown in the three different scenarios, the calculation of the amount of CO 2 storage potential for a formation can vary significantly depending upon the information known. The workflow presented (Fig. 1) gives storage efficiency values to use if net-to-gross area (target area) or net-to-gross area and thickness values are known for a formation [7]. These updated efficiency values will increase the accuracy of storage estimates as more appropriate efficiency values can be used based upon the amount of information known. Misapplication of storage efficiency values could lead to misinterpretation of the formation and may limit future studies if a formation is regarded as being low in storage potential. Conversely, the opposite could be true where a formation is seen as having high potential for storage when, in fact, the incorrect efficiency values were applied. Scenarios were presented to show the applicability of using the developed workflow to properly assess the correct efficiency term for a formation based on the amount of data available. Although each scenario has a storage mass calculated, each differs because of the amount of data and the geologic knowledge of the reservoir. Scenario 1 represents a typical quick-look assessment using the bare minimum of data. These values can be found with relatively little effort in publications by the state geological surveys or peer-reviewed journals. Because no net-togross terms are known, the efficiency factors are small and take a conservative approach to account for the unknown data, resulting in higher overall uncertainty in the assessment.

7 Wesley A. Peck et al. / Energy Procedia 63 ( 2014 ) Scenario 2 represents the most thorough assessment, and in most cases like this, a 3-D geologic model is built to perform dynamic simulations to investigate formation injection scenarios. Although Scenario 2 is the more accurate assessment, it is imperative that the correct efficiency term is applied, as applying the wrong efficiency term would essentially eliminate the net-to-gross area twice, thereby drastically reducing the overall storage resource estimate unnecessarily. Scenario 3 represents a situation where the target area is known and an updated storage efficiency value needs to be used. This case is typical of many formation evaluations, and improves upon the basic assessment of Scenario 1. This scenario is important because investigators need to know that a formation meets the depth and salinity requirements, and if they do have that information, then they very likely know the extent of the formation that is amenable to the CO 2 injection. With this information available, then the updated storage efficiency values become important and allow for a more accurate storage calculation. While Scenarios 1 and 3 are more of a characterized storage resource estimate and Scenario 2 is an effective storage resource estimate (Fig. 2), all three scenarios have value and can give reasonable estimates, provided their limitations are taken into consideration. For example, Scenario 1 represents a screening-level assessment that may be useful for decision makers to determine where to invest further characterization efforts. Fig. 2.CO 2 storage classification framework [2].

8 5238 Wesley A. Peck et al. / Energy Procedia 63 ( 2014 ) Acknowledgements A special thanks to Angela Goodman of DOE for providing the intermediate storage efficiency values shown in Fig. 1. This material is based on work supported by DOE NETL under Award Nos. DE-FC26-05NT42592, DE- FE , and Cooperative Agreement No. DE-FC26-08NT This work was also prepared with the support of the IEA Greenhouse Gas R&D Programme under reference IEA/CON/13/208. References [1] U.S. Department of Energy National Energy Technology Laboratory, 2012, Carbon sequestration atlas of the United States and Canada (4th ed.). [2] IEA Greenhouse Gas R&D Programme, 2009, Development of storage coefficients for CO 2 storage in deep saline formations: 2009/12, October [3] U.S. Department of Energy National Energy Technology Laboratory, 2010, Carbon sequestration atlas of the United States and Canada (3rd ed.). [4] Carbon Sequestration Leadership Forum, 2005, A task force for review and development of standards with regards to storage capacity measurement Phase I: CSLF-T , 8 August 2005, 16 p., MeasurementTaskForce.pdf (accessed February 2013). [5] IEA Greenhouse Gas R&D Programme, 2014, CO 2 Storage Efficiency in Deep Saline Formations: A Comparison of Volumetric and Dynamic Storage Resource Estimation Methods: May 2014 (in press). [6] Ellett K, Zhang Q, Medina C, Rupp J, Wang G, Carr T, Uncertainty in regional-scale evaluation of CO 2 geologic storage resources comparison of the Illinois Basin (USA) and the Ordos Basin (China): Energy Procedia, 2013, v. 37, p [7] Goodman, Angela, U.S. Department of Energy Personal communication.